align: 修正技能播种路径:codex-home/skills/ 才是用户技能目录

2026-10-30 主机实测(临时 codex-home,生产未动)纠正:
先播 .agents/skills/ 时 skills/list 只返回 5 个系统技能,
3 个自有技能一个没出现;改播 skills/ 后返回 8 个,
3 个自有技能以 scope=user 出现。系统技能本身就在
skills/.system/ 下。
- skill_seed DEST_SUBDIR 由 .agents/skills 改为 skills,
  模块文档与 worker 注释同步修正。
- 测试期望路径同步改为 skills/。
- 对照表补实测结论与附带发现:Codex 自带系统技能
  imagegen,图片生成在 Codex 本体里不是无技能。

验证:skill_seed + chatgpt skills + websearch worker
共 14 项全过。主机探针文件已全部清理,生产未动。
This commit is contained in:
host committed 2026-10-30 01:16:49 +08:00
1 parent cf9d17f035
commit 378cddfb1e
4 files changed
+50 -39

No files matched your search

+29 -22
View File
@@ -4,16 +4,22 @@ Mirrors the websearch MCP seeding pattern in cloud/websearch.py,
but for skills, not MCP config:
- Source is cloud/skills/<name>/SKILL.md (the batch-1 skill pack).
- Destination is <codex-home>/.agents/skills/<name>/SKILL.md.
Why this path: Codex documents personal skills at
$HOME/.agents/skills and project skills at .agents/skills
(scanned from the working directory up to the repo root).
A Docker tenant task must not pollute its task repo/worktree
(seeding into the worktree would show up in the task diff),
so the tenant codex-home -- the per-project persistent home
that is mounted into the container at
/var/lib/codex-app-server/codex-home -- is the personal-level
analogue. This is the same channel websearch MCP seeding uses.
- Destination is <codex-home>/skills/<name>/SKILL.md.
Why this path: verified live on 2026-10-30 with the real
Codex binary (host.py skills-list against a temp codex-home
on the production host, production untouched). The first
attempt -- <codex-home>/.agents/skills/ -- was NOT discovered
(skills/list returned only the 5 system skills). Copying the
same SKILL.md files to <codex-home>/skills/ made them appear
as scope=user (5 system + 3 ours, count 8). System skills
themselves live at <codex-home>/skills/.system/, which is
the tell that this is the user-skills directory for a
CODEX_HOME. A Docker tenant task must not pollute its task
repo/worktree (seeding into the worktree would show up in
the task diff), so the tenant codex-home -- the per-project
persistent home that is mounted into the container at
/var/lib/codex-app-server/codex-home -- is the right channel.
This is the same channel websearch MCP seeding uses.
- Idempotent and non-destructive by design:
- a missing destination file is copied;
- an identical destination file is left alone;
@@ -24,16 +30,16 @@ but for skills, not MCP config:
- No config.toml change, no key, no network: skills are plain
instruction files, and the helper never writes secrets.
Honest boundary (see docs/chatgpt-codex-alignment.md): the
destination path follows the documented personal-skills pattern
mapped onto the tenant home. Whether a given Codex build picks
these skills up in a Docker tenant task still needs a live
verification task (skills/list + an explicit $live-info call),
the same kind of e2e probe websearch MCP got after its seeding
PR. Until that probe passes, do not claim the skills are live
in production tasks -- this helper only guarantees the files
land in the tenant home, idempotently, without clobbering
tenant edits.
Honest boundary (see docs/chatgpt-codex-alignment.md): discovery
itself is now live-verified (skills/list, scope=user, on a temp
home with the real Codex binary). What is NOT yet verified is
model invocation inside a real Docker tenant task (an explicit
$live-info call that actually runs and returns a result, and
whether the relay model honours it). Until a task-level probe
passes, do not claim the skills are live in production tasks --
this helper only guarantees the files land in the tenant home,
idempotently, without clobbering tenant edits, and that
skills/list sees them there.
"""
import os
@@ -42,8 +48,9 @@ from pathlib import Path
# cloud/skills/<name>/SKILL.md, next to this module.
SKILLS_DIR = Path(__file__).resolve().parent / "skills"
# Personal-level skills inside a codex-home.
DEST_SUBDIR = Path(".agents") / "skills"
# User-level skills inside a codex-home (live-verified 2026-10-30:
# .agents/skills inside a codex-home is NOT discovered).
DEST_SUBDIR = Path("skills")
def skill_source_dir(skills_dir=None):
+3 -3
View File
@@ -56,7 +56,7 @@ class SkillSeedHelperTestCase(unittest.TestCase):
home = os.path.join(self.tmp.name, "home")
src = self._src()
ensure_codex_home_skills(home, src)
dest = os.path.join(home, ".agents", "skills",
dest = os.path.join(home, "skills",
"live-info", "SKILL.md")
open(dest, "w").write("tenant edited version\n")
out = ensure_codex_home_skills(home, src)
@@ -98,7 +98,7 @@ class WorkerSkillSeedTestCase(unittest.TestCase):
return Worker(self.db_path, "/fake/host.py", **kwargs)
def _skill_path(self, dirs, name):
return os.path.join(dirs["codex_home"], ".agents", "skills",
return os.path.join(dirs["codex_home"], "skills",
name, "SKILL.md")
def test_flag_off_is_old_behaviour(self):
@@ -106,7 +106,7 @@ class WorkerSkillSeedTestCase(unittest.TestCase):
self.assertFalse(worker.seed_skills)
dirs = worker._tenant_dirs("projA")
self.assertFalse(os.path.exists(
os.path.join(dirs["codex_home"], ".agents", "skills")))
os.path.join(dirs["codex_home"], "skills")))
def test_flag_on_seeds_real_pack(self):
worker = self._worker(seed_skills=True)
+2 -2
View File
@@ -570,7 +570,7 @@ class Worker:
# config.toml + 可选的 websearch MCP)。开启后
# _tenant_dirs 在 websearch 播种之后调用 skill_seed
# 的播种助手,把 cloud/skills/ 的 SKILL.md 拷进租户
# codex-home/.agents/skills/(个人级技能路径,见
# codex-home/skills/(用户级技能路径,实测见
# skill_seed.py 的路径说明)。老租户下一次任务自动
# 合并;租户改过的技能文件绝不覆盖,只上报 differs。
self.seed_skills = bool(seed_skills)
@@ -5702,7 +5702,7 @@ def main(argv=None):
parser.add_argument("--seed-skills", action="store_true",
help="租户 codex-home 自动播种对齐技能包:"
"_tenant_dirs 把 cloud/skills/ 的 SKILL.md"
" 拷进租户 codex-home/.agents/skills/;"
" 拷进租户 codex-home/skills/;"
"默认关,老租户不受影响;租户改过的"
"技能文件绝不覆盖")
parser.add_argument("--task-timeout", type=float, default=1800,